Re: How to interpret the following Oracle trace file?

Brandon wrote:

> Hi there,
>
> Currently, we have just upgraded our AIX OS to Version 4.3, running an
> Oracle 7.3.3 database, and we are having core dump file many times in
> a day.
> Has anyone out there face this problem before? Is upgrading of AIX
> will cause the core dump of oracle?
>
> In order to trace where the problem lies, we turn on the database
> tace, but
> we are having problem in interpreting the following trace file.
> I would very much appreciate if any of the oracle expert out there
> would
> like to help, thanks : )
>
> How do i interpret the following trace file? Is there a problem in it?
> And how to convert the "tim=69641995" into the actual date? Am I
> having
> any problem with the date configuration?
>
> Thanks in advance for taking your precious time in helping to looking
> into it : )
>
> rdgs,
> Brandon
>
> =====================
> PARSING IN CURSOR #10 len=49 dep=2 uid=11 oct=3 lid=11 tim=69641995
> hv=1258420735 ad='40c77e34'
> SELECT SWITCH_TYPE FROM LGW WHERE LGWIDX = :b1
> END OF STMT
> PARSE #10:c=0,e=0,p=0,cr=0,cu=0,mis=0,r=0,dep=2,og=4,tim=69641995
> EXEC #10:c=0,e=0,p=0,cr=0,cu=0,mis=0,r=0,dep=2,og=4,tim=69641995
> FETCH #10:c=0,e=0,p=0,cr=1,cu=2,mis=0,r=1,dep=2,og=4,tim=69641995
> EXEC #9:c=0,e=0,p=0,cr=1,cu=2,mis=0,r=1,dep=1,og=4,tim=69641995
> =====================
> PARSING IN CURSOR #11 len=51 dep=1 uid=11 oct=47 lid=11 tim=69641995
> hv=3947465601 ad='40c60058'
> begin :r:="NTMSRAW"."CBF"(:a1,:a2,:a3,:a4,:a5);end;
> END OF STMT
> PARSE #11:c=0,e=0,p=0,cr=0,cu=0,mis=0,r=0,dep=1,og=4,tim=69641995
> =====================
> PARSING IN CURSOR #12 len=49 dep=2 uid=11 oct=3 lid=11 tim=69641995
> hv=1258420735 ad='40c77e34'
> SELECT SWITCH_TYPE FROM LGW WHERE LGWIDX = :b1
> END OF STMT
> ..
> ..
> ..
> ..
>
> EXEC #6:c=0,e=0,p=0,cr=0,cu=0,mis=0,r=1,dep=1,og=4,tim=69645349
> EXEC #7:c=0,e=0,p=0,cr=0,cu=0,mis=0,r=1,dep=1,og=4,tim=69645349
> EXEC #8:c=0,e=0,p=0,cr=0,cu=0,mis=0,r=1,dep=1,og=4,tim=69645349
> EXEC #10:c=0,e=0,p=0,cr=0,cu=0,mis=0,r=1,dep=2,og=4,tim=69645349
> FETCH #10:c=0,e=0,p=0,cr=1,cu=2,mis=0,r=1
>
> Is there a problem when the trace file ends like last sentence above
> ??
>
> Thanks a lot for helping : )
>

Your trace file got cut of, the execution of curser 11.. is still missing. Change the init.ora max_dump_file_size to the number of OS blocks you dont want to exceed.

The trace file looks fine - so far. The time is a abitrary number - I think it is number of 1/100 seconds since instance start - it doesnt contain any date time value(s). Usable as to figure how long did this take, subtract begin time from end time => time for job (in 1/100 sec.)

rgds

/Svend Jensen Received on Fri May 24 2002 - 13:00:30 CDT